Errors in atmospheric tidal determination from surface pressure observations

It has previously been assumed that by taking a sufficient length of data the mean atmospheric tide at any tropical station could be determined to an accuracy of a few percentage points by using surface pressure observations. However, evidence is herein presented which shows that the inferred amplitude of the tide can be changed by over 20% just by changing the instrument used to measure atmospheric pressure. This has occurred both when one barometer has replaced another and when one barograph has replaced another at the same station. It is suggested that the main reason for the inaccuracy is that diurnal temperature variations cause systematic instrumental errors. There are important implications for the accuracy of all tidal determinations that have been made and for the reliability of tidal theory.